Untangling procedural and distributive justice.

Although there has been a significant increase in the amount of research that addresses both procedural and distributive justice, researchers continue to debate the relative importance of the two forms of justice in predicting employee attitudes (Sweeney & McFarlin, 1993). In an effort to help untangle this literature, this study investigates the degree to which procedural and distributive justice affect employee satisfaction with gainsharing. Most of the related justice research uses surveys distributed in one company. Employees are asked for their reactions to a variety of organizational programs (of which some are individually derived and others are group-based). As a result, researchers concluded that procedural justice was more important when an outcome was group-based, and distributive justice was more critical when the outcome was individual-based. The types of outcomes studied included group conflict, trust in management, organization commitment, pay satisfaction, and evaluations of supervisors (Alexander & Ruderman, 1987; Folger & Konovsky, 1989; McFarlin & Sweeney, 1992; Sweeney & McFarlin, 1993).

A separate body of work, conducted by Greenberg (1987b), suggested that the individual versus group context of an outcome was not the key factor in determining which type of justice was more important. Instead, Greenberg (1987b) suggested that favorableness of the outcome (high/low, positive/negative) was the determining factor. Procedural fairness was more important for an unfavorable or low outcome, and distributive fairness was more important for a favorable or high outcome.

In this study, I investigate the two interpretations of the relative importance of distributive and procedural justice with a research design that is different from what has been used in the past. As a result, this work contributes not only to the justice literature but also to what we know about gainsharing plans. In contrast to prior justice research, this study focuses only on a group-based context. Rather than studying employee attitudes toward both individual and group outcomes, this study addresses employee reactions to one type of outcome (a group-based, gainsharing plan). This extends the justice literature into a different context, and it allows me to focus on the effect of payout condition (high/low, favorable/unfavorable) within a group context.

The study was conducted in two companies, one with high or favorable bonus payouts and another with low and zero (or unfavorable) bonuses. Very few studies of unsuccessful gainsharing plans have been done; therefore, this particular application of the justice literature serves to build on our knowledge of gainsharing by studying both a successful and an unsuccessful (discontinued after three quarters' experience) plan.

The gainsharing context is particularly useful for studying the relative impacts of procedural and distributive justice, because the dominant interpretations speculate that the individual versus group context affects the relative importance of these two types of justice (Lind & Tyler, 1988). Some of the strongest findings come from pay-related studies; however, prior research has not varied the pay context. In fact, McFarlin and Sweeney (1992) suggested that future research should investigate employee reactions to outcomes other than those studied to date. The study reported in this article responds to their call for new research designs by examining distributive and procedural justice in a group-based context and focusing on employee reactions to gainsharing programs.

GAINSHARING

Gainsharing plans were first used in the 1930s by Joseph Scanlon as a way to turn around the financially troubled Empire Steel and Tin Plate Company. His use of a bonus formula - including only financial components that employees could affect (e.g., payroll costs and net sales value of production) and suggestion committees that gave employees a way to improve the outcomes included in the formula - became popular as a way to enhance performance in other organizations (O' Dell, 1981). Today, interest in gainsharing continues to increase (Towers Perrin, 1997). However, bonus formulas are being customized to reflect unique organizational goals. Whereas traditional gainsharing plans were implemented primarily in manufacturing environments, which had stable financial histories that allowed for clear establishment of the gainsharing bonus criteria, today's gainsharing plans also are being used in hospitals, banks, restaurants, and other service organizations, in which measurement is more difficult (Markham, Scott, & Little, 1992). As a result, in addition to diversity in settings, gainsharing formulas have been altered to include components for customer service, safety, and quality (Welbourne & Gomez-Mejia, 1995).

Gainsharing plans are implemented to increase employee effort and to encourage and reward behaviors identified as critical for organizational effectiveness (Welbourne & Gomez-Mejia, 1995). Gainsharing does this by attempting to create a sense of fairness among employees by distributing the "gains" equally between the employer and employees and then distributing the employee portion among all participants in the gainsharing unit (see Graham-Moore & Ross, 1990, for a description of gainsharing formulas and process). The goal of "fairness" in gainsharing design (as documented in many case studies) has been emphasized by managers and consultants involved with the implementation of these programs.

Although gainsharing has emerged independent of academic theory (its founder was Joseph Scanlon - a boxer, cost accountant, and union leader), the role of fairness expressed by practitioners is consistent with predictions based on justice theories. Equity concepts (as defined by Adams, 1965, and Homans, 1960) have spurred volumes of research and have been used to understand how allocation decisions are made within organizations and how outcome distribution affects productivity levels, satisfaction among workers, or both (Deutsch, 1975; Leventhal, 1976). Early research on equity theory focused on distributive justice, which is defined as the fairness perceptions of consequences or outcomes (such as pay) that result from various decision processes. During the 1970s, the justice literature was expanded to include the concept that the fairness perceptions of procedures had an independent role to play in understanding employee reactions to decisions and allocations (Lind & Tyler, 1988). Thus, a body of research on procedural fairness, which investigated perceptions of the decision processes themselves, began to develop.

PROCEDURAL AND DISTRIBUTIVE JUSTICE IMPLICATIONS

The original work in procedural justice was grounded in legal research (Thibaut & Walker, 1975). Although this work addressed perceptions of legal systems and dispute resolution mechanisms, it eventually was extended to other settings such as police contact (Tyler & Folger, 1980), evaluations of politicians and teachers (Tyler & Caine, 1981), dispute resolution in organizations (Sheppard, 1985), and performance appraisal (Greenberg, 1986). The results from this research supported the original finding that procedural justice was an important factor for understanding individual attitudes, and it was independent of …